Neural Control of Unknown Non-Affine Systems by State and Output Feedback

Wei He, Dan Deng, Zhong-Liang Tang

Abstract: In this paper, we consider the unknown non-affine nonlinear systems with the input deadzone and external disturbance. To solve the deadzone effects, a Radial Basis Function Neural Network (RBFNN) is introduced to approximate the deadzone effects and improve the robustness of the system. Moreover, we estimate the disturbance in another RBFNN. Based on the Lyapunov stability analysis and the support of high-gain observer, the adaptive neural network control is developed via both full-state feedback and output feedback. Simulation studies are proposed to illustrate the control performance. The semiglobally uniform boundedness is guaranteed in the closed-loop systems under the proposed control.